void MapAlignmentAlgorithmIdentification::alignPeptideIdentifications(
vector<vector<PeptideIdentification> > & maps,
vector<TransformationDescription> & transformations)
{
checkParameters_(maps.size());
startProgress(0, 3, "aligning peptide identifications");
if (reference_index_) // reference is one of the input files
{
SeqToList rt_data;
getRetentionTimes_(maps[reference_index_ - 1], rt_data);
computeMedians_(rt_data, reference_, true);
}
// one set of RT data for each input map, except reference:
vector<SeqToList> rt_data(maps.size() - bool(reference_index_));
for (Size i = 0, j = 0; i < maps.size(); ++i)
{
if (i == reference_index_ - 1)
continue; // skip reference map, if any
getRetentionTimes_(maps[i], rt_data[j++]);
}
setProgress(1);
computeTransformations_(rt_data, transformations, true);
setProgress(2);
// transformPeptideIdentifications(maps, transformations);
setProgress(3);
endProgress();
}
void MapAlignmentAlgorithmIdentification::getRetentionTimes_(
MapType & features, SeqToList & rt_data)
{
bool use_feature_rt = param_.getValue("use_feature_rt").toBool();
for (typename MapType::Iterator feat_it = features.begin();
feat_it != features.end(); ++feat_it)
{
if (use_feature_rt)
{
// find the peptide ID closest in RT to the feature centroid:
String sequence;
DoubleReal rt_distance = numeric_limits<DoubleReal>::max();
bool any_good_hit = false;
for (vector<PeptideIdentification>::iterator pep_it =
feat_it->getPeptideIdentifications().begin(); pep_it !=
feat_it->getPeptideIdentifications().end(); ++pep_it)
{
if (hasGoodHit_(*pep_it))
{
any_good_hit = true;
DoubleReal current_distance =
abs(double(pep_it->getMetaValue("RT")) - feat_it->getRT());
if (current_distance < rt_distance)
{
sequence = pep_it->getHits()[0].getSequence().toString();
rt_distance = current_distance;
}
}
}
if (any_good_hit)
rt_data[sequence] << feat_it->getRT();
}
else
{
getRetentionTimes_(feat_it->getPeptideIdentifications(), rt_data);
}
}
if (!use_feature_rt && param_.getValue("use_unassigned_peptides").toBool())
{
getRetentionTimes_(features.getUnassignedPeptideIdentifications(),
rt_data);
}
// remove duplicates (can occur if a peptide ID was assigned to several
// features due to overlap or annotation tolerance):
for (SeqToList::iterator rt_it = rt_data.begin();
rt_it != rt_data.end(); ++rt_it)
{
DoubleList & rt_values = rt_it->second;
sort(rt_values.begin(), rt_values.end());
DoubleList::iterator it = unique(rt_values.begin(), rt_values.end());
rt_values.resize(it - rt_values.begin());
}
}
void MapAlignmentAlgorithmIdentification::setReference(Size reference_index,
const String & reference_file)
{
reference_.clear();
reference_index_ = reference_index;
// reference is one of the input files, or no reference given:
if (reference_index_ || reference_file.empty())
return;
// reference is external file:
LOG_DEBUG << "Extracting reference RT data..." << endl;
SeqToList rt_data;
bool sorted = true;
FileTypes::Type filetype = FileHandler::getType(reference_file);
if (filetype == FileTypes::MZML)
{
MSExperiment<> experiment;
MzMLFile().load(reference_file, experiment);
getRetentionTimes_(experiment, rt_data);
sorted = false;
}
else if (filetype == FileTypes::FEATUREXML)
{
FeatureMap<> features;
FeatureXMLFile().load(reference_file, features);
getRetentionTimes_(features, rt_data);
}
else if (filetype == FileTypes::CONSENSUSXML)
{
ConsensusMap features;
ConsensusXMLFile().load(reference_file, features);
getRetentionTimes_(features, rt_data);
}
else if (filetype == FileTypes::IDXML)
{
vector<ProteinIdentification> proteins;
vector<PeptideIdentification> peptides;
IdXMLFile().load(reference_file, proteins, peptides);
getRetentionTimes_(peptides, rt_data);
}
computeMedians_(rt_data, reference_, sorted);
if (reference_.empty())
{
throw Exception::MissingInformation(
__FILE__, __LINE__, __PRETTY_FUNCTION__,
"Could not extract retention time information from the reference file");
}
}
// lists of peptide hits in "maps" will be sorted
void MapAlignmentAlgorithmIdentification::getRetentionTimes_(
MSExperiment<> & experiment, SeqToList & rt_data)
{
for (MSExperiment<>::Iterator exp_it = experiment.begin();
exp_it != experiment.end(); ++exp_it)
{
getRetentionTimes_(exp_it->getPeptideIdentifications(), rt_data);
}
// duplicates should not be possible -> no need to remove them
}
// lists of peptide hits in "maps" will be sorted
bool MapAlignmentAlgorithmIdentification::getRetentionTimes_(
PeakMap& experiment, SeqToList& rt_data)
{
for (PeakMap::Iterator exp_it = experiment.begin();
exp_it != experiment.end(); ++exp_it)
{
getRetentionTimes_(exp_it->getPeptideIdentifications(), rt_data);
}
// duplicate annotations should not be possible -> no need to remove them
return false;
}